ClawHub
Back to skill

Security audit

🎨 Text-to-Image Free

Security checks across malware telemetry and agentic risk

Overview

This is a straightforward image-generation skill that sends prompts to Pollinations.ai and saves the returned image locally, with no evidence of hidden persistence, credential theft, or destructive behavior.

Install only if you are comfortable sending image prompts to Pollinations.ai. Do not put secrets, personal information, regulated data, unreleased business material, or confidential project details in prompts, and be cautious with ambiguous requests that include words like generate or draw.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list includes very broad phrases like "generate," "draw," and "make a picture," which can match ordinary conversation and cause the skill to activate when the user did not explicitly intend to use this external image-generation tool. Because activation leads to sending user-provided content to a third-party API, unintended triggering increases the risk of privacy leakage and unexpected external actions.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The Chinese trigger keywords include broad everyday terms such as "画" and "生成图片," which may be interpreted in normal conversation rather than as a deliberate request to invoke this skill. In this skill, accidental activation is more concerning because it can forward user text to an external service without a clear consent step.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly instructs the agent to send the user's prompt to Pollinations.ai via curl, but it does not clearly warn that prompt contents are disclosed to a third-party service. Users may include sensitive or confidential information in prompts, so the absence of a prominent privacy notice can lead to unintended data exposure outside the local agent environment.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The Chinese documentation describes forwarding prompts to Pollinations.ai but does not prominently disclose that user input leaves the agent and is processed by a third-party API. This creates the same privacy risk for Chinese-speaking users, who may reasonably assume the action is local unless explicitly warned otherwise.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.